共 81 条
QSPR molecular approach for representation/prediction of very large vapor pressure dataset
被引:36
作者:
Gharagheizi, Farhad
[2
]
Eslamimanesh, Ali
[1
]
Ilani-Kashkouli, Poorandokht
[2
]
Mohammadi, Amir H.
[1
,3
]
Richon, Dominique
[1
,3
]
机构:
[1] CEP TEP Ctr Energet & Proc, MINES ParisTech, F-77305 Fontainebleau, France
[2] Islamic Azad Univ, Buinzahra Branch, Dept Chem Engn, Buinzahra, Iran
[3] Univ KwaZulu Natal, Sch Chem Engn, Thermodynam Res Unit, ZA-4041 Durban, South Africa
关键词:
Vapor pressure;
Quantitative structure-property relationship;
Computational chemistry;
Optimization;
Thermodynamics process;
Phase equilibria;
STRUCTURE-PROPERTY RELATIONSHIPS;
SUPERCRITICAL CARBON-DIOXIDE;
INDUSTRIAL SOLID COMPOUNDS;
PENG-ROBINSON EQUATION;
CRITICAL-TEMPERATURES;
CHEMICAL-PROPERTIES;
ORGANIC-COMPOUNDS;
CUBIC EQUATIONS;
THERMAL DATA;
OF-STATE;
D O I:
10.1016/j.ces.2012.03.033
中图分类号:
TQ [化学工业];
学科分类号:
0817 ;
摘要:
Reliable estimation of vapor pressure is of great significance for chemical industry. In this communication, the capability of the Quantitative StructureProperty Relationship (QSPR) technique is studied to represent/predict the vapor pressure of pure chemical compounds from about 55 to around 3040 K. Around 45,000 vapor pressure values belonging to about 1500 chemical compounds (mostly organic ones) at different temperatures are treated in order to present a comprehensive, reliable, and predictive model. The sequential search mathematical method has been observed to be the only variable search method capable of selection of appropriate model parameters (molecular descriptors) regarding this extremely large data set. To develop the final model, a three-layer artificial neural network is optimized using the LevenbergMarquardt (LM) optimization strategy. Through the developed QSPR model, the absolute average relative deviation of the represented/predicted properties from the applied data is about 7% and squared correlation coefficient is 0.990. In addition, the outliers of the model are identified using the Leverage Value Statistics method. (c) 2012 Elsevier Ltd. All rights reserved.
引用
收藏
页码:99 / 107
页数:9
相关论文